nep-ecm New Economics Papers
on Econometrics
Issue of 2009‒09‒05
nine papers chosen by
Sune Karlsson
Orebro University

  1. System GMM Estimation With A Small Sample By Marcelo Soto
  2. Dominating estimators for the global minimum variance portfolio By Frahm, Gabriel; Memmel, Christoph
  3. Copula-based bivariate binary response models By Rainer Winkelmann
  4. Generalized Impulse Response Analysis: General or Extreme? By Hyeongwoo, Kim
  5. Time dynamic and hierarchical dependence modelling of an aggregated portfolio of trading books: a multivariate nonparametric approach By Gaisser, Sandra; Memmel, Christoph; Schmidt, Rafael; Wehn, Carsten
  6. Testing unilateral and bilateral link formation By Margherita Comola; Marcel Fafchamps
  7. Impact identification strategies for evaluating business incentive programs. By Bondonio, Daniele
  8. Correcting for Survival Effects in Cross Section Wage Equations Using NBA Data By Peter A. Groothuis; James Richard Hill
  9. Measuring Inequality Using Censored Data: A Multiple Imputation Approach By Stephen Jenkins; Richard Burkhauser; Shuaizhang Feng; Jeff Larrimore

  1. By: Marcelo Soto
    Abstract: Properties of GMM estimators for panel data, which have become very popular in the empirical economic growth literature, are not well known when the number of individuals is small. This paper analyses through Monte Carlo simulations the properties of various GMM and other estimators when the number of individuals is the one typically available in country growth studies. It is found that, provided that some persistency is present in the series, the system GMM estimator has a lower bias and higher efficiency than all the other estimators analysed, including the standard first-differences GMM estimator.
    Keywords: Economic Growth, System GMM estimation, Monte Carlo Simulations
    JEL: C15 C33 O11
    Date: 2009–09–01
  2. By: Frahm, Gabriel; Memmel, Christoph
    Abstract: Two shrinkage estimators for the global minimum variance portfolio that dominate the traditional estimator with respect to the out-of-sample variance of the portfolio return are derived. The presented results hold for any number of observations n >= d 2 and number of assets d >= 4. The small-sample properties of the shrinkage estimators and also their large-sample properties for fixed d but n -> infinity as well as n,d -> infinity but n/d -> q <= infinity are investigated. Further, a small-sample test for the question whether it is better to completely ignore time series information in favor of naive diversification is presented.
    Keywords: Covariance matrix estimation,global minimum variance portfolio,James-Stein estimation,naive diversification,shrinkage estimator
    JEL: C13 G11
    Date: 2009
  3. By: Rainer Winkelmann (Socioeconomic Institute, University of Zurich)
    Abstract: The bivariate probit model is frequently used for estimating the effect of an endogenous binary regressor on a binary outcome variable. This paper discusses simple modifications that maintain the probit assumption for the marginal distributions while introducing non-normal dependence among the two variables using copulas. Simulation results and evidence from two applications, one on the effect of insurance status on ambulatory expenditure and one on the effect of completing high school on subsequent unemployment, show that these modified bivariate probit models work well in practice, and that they provide a viable and simple alternative to the standard bivariate probit approach.
    Keywords: Bivariate probit, binary endogenous regressor, Frank copula, Clayton copula
    JEL: C23
    Date: 2009–08
  4. By: Hyeongwoo, Kim
    Abstract: This note discusses a pitfall of using the generalized impulse response function (GIRF) in vector autoregressive (VAR) models (Pesaran and Shin, 1998). The GIRF is general because it is invariant to the ordering of the variables in the VAR. The GIRF, in fact, is extreme because it yields a set of response functions that are based on extreme identifying assumptions that contradict each other, unless the covariance matrix is diagonal. With an empirical example, the present note demonstrates that the GIRF may yield quite misleading economic inferences.
    Keywords: Generalized Impulse Response Function; Orthogonalized Impulse Response Function; Vector Autoregressive Models
    JEL: C32 C13 C51
    Date: 2009–04
  5. By: Gaisser, Sandra; Memmel, Christoph; Schmidt, Rafael; Wehn, Carsten
    Abstract: From a banking supervisory perspective, this paper analyses aspects of market risk of an aggregated trading portfolio comprised of the trading books of 11 German banks with a regulatory approved internal market risk model. Based on real, clean profit and loss data and Value-at-Risk estimates of the 11 banks, the paper specifically models and analyzes the portfolio's dependence and diversification structure, indispensable for financial stability studies. The high sensitivity of market risk measurements with respect to dependence structure of the underlying portfolio is nowadays a well-known fact. However, only few techniqques for high-dimensional and hierarchical dependence analysis have been proposed and studied in the financial literature so far. One reason is certainly the increasing complexity of the statistical theory, which is commonly referred to as the curse of high-dimensionality. The present paper develops and applies multidimensional (asymptotic) test statistics based on the copula theory with the aim of detecting significant long-term level changes in the supervisory portfolio's dependence over time. Furthremore, a statistical hyphothesis test is proposed to identify the distinct contributions of sub-portfolios towards the overall dependence level in ahiercharchical manner. The utilized techniques are distribution-free and, in particulaar, are invariant with respect to the maarginaal return distributions.
    Keywords: Multivariate dependence modelling,multivariate Spearman's rho,time-varying copula,asymptotic test theory,hierarchical testing,control chart theory
    JEL: C12 C13 C14
    Date: 2009
  6. By: Margherita Comola; Marcel Fafchamps
    Abstract: The literature has shown that network architecture depends crucially on whether links are formed unilaterally or bilaterally, that is, on whether the consent of both nodes is required for a link to be formed. We propose a test of whether network data is best seen as an actual link or willingness to link and, in the latter case, whether this link is generated by an unilateral or bilateral link formation process. We illustrate this test using survey answers to a risk-sharing question in Tanzania. We find that the bilateral link formation model fits the data better than the unilateral model, but the data are best interpreted as willingness to link rather than an actual link. We then expand the model to include self-censoring and find that models with self-censoring fit the data best.
    Date: 2009
  7. By: Bondonio, Daniele
    Abstract: Although business incentive programs of different forms have been the bulk of local economic development policies in many industrialized countries for more than the last three decades, evaluating their impact on employment or local economic growth outcomes remains a challenging task due to the persisting lack of randomized experiments and the presence of many confounding factors which affect firms and economic growth outcomes. Moreover, much of the recent advancements in the statistical program evaluation methodology applicable to non-experimental settings do not make any direct reference to the specificities posed by business incentive policies. This paper aims at offering some clear guidance on how to choose the appropriate focus of the evaluation, the policy relevant evaluation parameters and empirical impact identification strategies when applying statistical methods attempting to estimate how much of the different outcomes between treatment and control groups are attributable to the program/s being evaluated. Each methodological option discussed in the paper is linked to the different features of commonly implemented US and EU policies and to whether or not the analysis focuses on outcomes recorded at a firm-level or at the level of the geographic areas in which the assisted firms are located.
    Keywords: Impact evaluation, Business Incentives Policies; Comparison-group designs, Identification strategies
    JEL: C40 C80 H81
    Date: 2009–07
  8. By: Peter A. Groothuis; James Richard Hill
    Abstract: Cross sectional employment data is not random. Individuals who survive to a longer level of tenure tend to have a higher level of productivity than those who exit earlier. This result suggests that in cross sectional data high productivity workers are over-sampled at high levels of tenure. In wage equations using cross sectional data, results could be biased from the over sampling of high productive workers at long levels of tenure. This survival effect in cross sectional data could possibly bias the coefficient on tenure upwards. We explore techniques to correct for survival bias using a panel study of National Basketball Association players. In particular we focus on a modified Heckman selectivity bias procedure using duration models to correct for survival bias. Key Words:
    Date: 2009
  9. By: Stephen Jenkins; Richard Burkhauser; Shuaizhang Feng; Jeff Larrimore
    Abstract: To measure income inequality with right censored (topcoded) data, we propose multiple imputation for censored observations using draws from Generalized Beta of the Second Kind distributions to provide partially synthetic datasets analyzed using complete data methods. Estimation and inference uses Reiter’s (Survey Methodology 2003) formulae. Using Current Population Survey (CPS) internal data, we find few statistically significant differences in income inequality for pairs of years between 1995 and 2004. We also show that using CPS public use data with cell mean imputations may lead to incorrect inferences about inequality differences. Multiply-imputed public use data provide an intermediate solution.
    Keywords: Income Inequality, Topcoding, Partially Synthetic Data, CPS, Current Population Survey, Generalized Beta of the Second Kind distribution
    JEL: D31 C46 C81
    Date: 2009–04

This nep-ecm issue is ©2009 by Sune Karlsson. It is provided as is without any express or implied warranty. It may be freely redistributed in whole or in part for any purpose. If distributed in part, please include this notice.
General information on the NEP project can be found at For comments please write to the director of NEP, Marco Novarese at <>. Put “NEP” in the subject, otherwise your mail may be rejected.
NEP’s infrastructure is sponsored by the School of Economics and Finance of Massey University in New Zealand.